Atypical Membrane Topology and Heteromeric Function of Drosophila Odorant Receptors In Vivo

A mosquito smells you before it sees you. From meters away, it reads a chemical signature — your breath, your skin, the particular mix of compounds that mark you as a warm-blooded target. That capacity to detect and pursue a host is mediated by odorant receptors packed into tiny sensory hairs on the insect's antennae. For years, researchers assumed those receptors worked roughly the same way mammalian smell receptors do — with the same basic architecture and molecular logic. A paper by Benton, Sachse, Michnick, and Vosshall demonstrated that assumption was wrong in a way that changes everything. Insect olfactory receptors are built differently, they work differently, and that difference turns out to be exactly what makes them a promising target for the next generation of insect repellents. Here is what was known going in. Each olfactory sensory neuron in Drosophila expresses two odorant receptor proteins. One is a conventional ligand-binding receptor — divergent from neuron to neuron, which is how the fly builds its olfactory map. The other is a receptor called OR83b, which is nearly identical across all these neurons. OR83b is conserved. It is everywhere. Flies without it are essentially anosmic — unable to smell. In Or83b null mutants, conventional receptors like OR22a deteriorate badly, detectable only at trace levels, and never in the right location. OR83b was clearly essential, but how it worked remained a genuine mystery. Part of the confusion came from analogy. Mammalian odorant receptors are G protein-coupled receptors, or GPCRs — seven-pass transmembrane proteins that sit in the cell membrane, bind odor molecules on one face, and trigger signaling cascades through G proteins on the other. Because insect odorant receptor sequences also predicted seven transmembrane domains, the field assumed insects probably used a similar architecture. Benton and colleagues suspected otherwise, and their experiments proved it. The first thing they established was that OR83b acts as a universal trafficking partner. When OR83b is absent, conventional odorant receptors never make it to where they need to go. Olfactory signaling occurs in the sensory cilia — tiny hair-like projections at the tips of the neuron's dendrites. Think of them as the antenna of the antenna. In Or83b mutants, conventional receptors stall in the cell body, failing to reach the cilia, with partial overlap with endoplasmic reticulum markers in about twenty percent of neurons. The endoplasmic reticulum is the first stop in the endomembrane system — the cell's internal trafficking highway — so these receptors weren't just misrouted; they were trapped near the very beginning of the journey. The team used a system called TARGET, which allows them to turn gene expression on or off in specific neurons at specific times. When they switched OR83b back on in adult neurons — after development had already finished — OR22a ciliary localization was restored. When they switched OR83b off, OR22a levels declined progressively and disappeared from cilia within days. The requirement for OR83b was not developmental. It was continuous. The neuron needed OR83b in the same cell throughout adult life to keep conventional odorant receptors in the right place. When they expressed a conventional odorant receptor, a GFP-tagged version of OR43a, in neurons that don't normally express any odorant receptor, it failed to reach cilia on its own. Co-express OR83b alongside it, and both proteins localize correctly. OR83b is the escort. Without it, the cargo goes nowhere. To prove the two proteins actually associate physically — rather than simply requiring each other at a distance — Benton and colleagues deployed a technique called split-YFP protein-fragment complementation. The logic is elegant: take a yellow fluorescent protein and cut it in half. Neither half glows on its own. But if the two halves are brought within roughly eighty angstroms of each other — held together by interacting proteins fused to each fragment — fluorescence reconstitutes. The glow indicates that the proteins are touching, and its location indicates where. When they fused complementary YFP fragments to two copies of OR83b and expressed them together, fluorescence appeared in the cell body and sensory dendrites — OR83b forms homomers, dimers with itself. They then paired one YFP fragment on OR43a with the other on OR83b. Robust fluorescence concentrated in the sensory cilia, with discrete puncta near the nucleus and in the inner dendrite. OR43a and OR83b are physically associated at the very site of odor detection. Controls using the gustatory receptor GR21a instead of OR83b produced only faint background signal, confirming the interaction is specific. Physical association is necessary but not sufficient. The team needed to show the complex actually signals. To do that, they used G-CaMP, a genetically encoded calcium reporter that lights up when a neuron fires. They expressed GFP-tagged OR43a together with OR83b in neurons that normally respond only to carbon dioxide — neurons with a completely different molecular identity. Those neurons, reconstituted with the OR43a and OR83b complex, now responded to cyclohexanol, cyclohexanone, hexanol, benzaldehyde, isoamyl acetate, and geranyl acetate — all known OR43a ligands — but not to control odors like octanol or linalool. The responses were statistically significant, with a p-value below 0.05, with four animals per genotype per stimulus. The OR and OR83b complex is not just necessary for signaling. It is sufficient to confer odorant responsiveness on a neuron that had none. Now comes the structural surprise. While all this was being established, Benton and colleagues were also asking a more fundamental question: how are these proteins oriented in the membrane? In a G protein-coupled receptor, the N-terminus — the protein's beginning — sits outside the cell. That extracellular face is where ligand binding typically occurs. When the team analyzed Drosophila odorant receptors, the topology was flipped. The N-terminus is inside, in the cytoplasm. The most conserved loops face inward. What would be the extracellular second loop in a G protein-coupled receptor faces outward here, but the rest of the architecture is inverted. They proved this with three independent approaches. In cultured Drosophila cells, beta-galactosidase fusion constructs confirmed an intracellular N-terminus — the enzyme was only active when on the cytoplasmic side. In living olfactory neurons, a split-YFP topology sensor placed on cytoplasmic-facing fragments reconstituted fluorescence in cilia and cell bodies, confirming the orientation in the actual sensory context. And immunoelectron microscopy — gold-labeled antibodies visualized under an electron microscope — placed eighty-seven point five percent of four hundred seventy-one gold particles outside the ciliary membrane when targeting the second extracellular loop, confirming that loop's outside-facing position. The topology is real, it is consistent across multiple methods, and it applies to both OR83b and conventional odorant receptors. The functional punchline is that this inverted topology is inseparable from the mechanism of heteromerization. The cytoplasmic loops that face inward are the ones that mediate OR and OR83b association. Using chimeric receptors and yeast two-hybrid assays, Benton and colleagues mapped the interaction to intracellular loop three of OR43a binding to intracellular loop three of OR83b. That same OR83b region interacts with OR22a's equivalent domain, but not with GR21a's — again confirming specificity. The structural oddity and the functional mechanism are not separate phenomena. They are the same thing. Zooming out, the implication becomes striking. OR83b is nearly identical across insects — not just Drosophila, but mosquitoes, moths, and beetles. OR83b orthologues from diverse insects can functionally substitute for the Drosophila version. This is an insect-wide architecture. And because it is structurally unrelated to mammalian G protein-coupled receptor-family odorant receptors — with inverted topology, obligate heteromeric complex, and universally required co-receptor — it represents what Benton and colleagues call an insect-specific solution to odor recognition. That specificity is the practical opening. A compound that disrupts the OR and OR83b interface, or jams the complex's interaction with the ciliary trafficking pathway, would have no mammalian counterpart to accidentally hit. It would be targeting a lock that exists in mosquitoes and not in us. The behaviors at stake are the ones that make insect-borne disease possible — the chemosensory host-seeking that brings a malaria-carrying mosquito to a human arm. The very features that made insect olfaction strange and hard to interpret — the backwards topology, the obligate co-receptor, the conserved heteromeric architecture — are precisely what makes it attractive as a target. What looked like a puzzle turned out to be a door. This lecture was created by ennepō. Go to https://ennepo.ai to Discover, Create and Follow the latest research in your field. Read when you can. Listen when you want to.